機械学習を用いたHILSシステムによる車体上下動評価

抄録

In recent years, model-based development such as the HILS system has been used to save time and cost in the development of complex systems such as automobiles. HILS system is a hybrid method with vehicle dynamics analysis and bench test. The author's group developed a testing machine to evaluate the characteristics of tire-suspension systems and constructed a HILS system. In the HILS system of this study, the measured damper force was used as the input to the analysis model, and the actuator of the testing machine was controlled in order to reproduce the suspension displacement in the analysis model. In this study, two kinds of control method were discussed: control with a linear model and IDCS with machine learning model. The reproducibility of the vehicle body vertical movement is evaluated by comparing the suspension strokes of the analysis model and the testing machine.